Who Mobilizes? Participatory Democracy in Chávez’s ... · This article advances this discussion by studying popular mobiliza-tion under the government of Hugo Chávez in Venezuela,

Who Mobilizes?Participatory Democracy in

Chávez’s Bolivarian Revolution

Kirk A. Hawkins

ABSTRACT

This article assesses popular mobilization under the Chávez gov-ernment’s participatory initiatives in Venezuela using data from theAmericasBarometer survey of 2007. This is the first study of the so-called Bolivarian initiatives using nationally representative, individ-ual-level data. The results provide a mixed assessment. Most of thegovernment’s programs invite participation from less active seg-ments of society, such as women, the poor, and the less educated,and participation in some programs is quite high. However, muchof this participation clusters within a narrow group of activists, anda disproportionate number of participants are Chávez supporters.This partisan bias probably reflects self-screening by Venezuelanswho accept Chávez’s radical populist discourse and leftist ideology,rather than vote buying or other forms of open conditionality. Thus,the Venezuelan case suggests some optimism for proponents of par-ticipatory democracy, but also the need to be more attuned to itspractical political limits.

During the past decade, leftist governments with participatory dem-ocratic agendas have come to power in many Latin American coun-

tries, implementing institutional reforms at the local and, increasingly,the national level. This trend has generated a scholarly literature assess-ing the nature of participation in these initiatives; that is, whether theyembody effective attempts at participatory forms of democracy thatmobilize and empower inactive segments of society (Goldfrank 2007;Wampler 2007a).

This article advances this discussion by studying popular mobiliza-tion under the government of Hugo Chávez in Venezuela, referred tohere as the Bolivarian Revolution or Chavismo. Under Chávez’s ener-getic efforts, and aided by an enormous surge in oil revenues,Venezuela’s government has implemented an incredible variety of pro-grams designed to create what it calls a “participatory and protagonisticdemocracy” (República Bolivariana de Venezuela 2001). While Chav-ismo clearly has attributes that make it sui generis, such as the charis-matic persona of Chávez, its struggle to implement participatory democ-racy is the most ambitious nationwide effort to date, and it helps makethe movement an important reference point for scholars and activists.

© 2010 University of Miami

This article uses a relatively new source of data, the AmericasBarometer 2006–7, a nationwide survey of 1,500 respondents conductedin August and September 2007 and sponsored by the Latin AmericanPublic Opinion Project (LAPOP) at Vanderbilt University. This is the firstsurvey in the series to be conducted in Venezuela, and it includes sev-eral sets of questions on popular participation in the erstwhile Bolivar-ian Circles; newer Bolivarian initiatives, such as the Communal Coun-cils; and the government’s broad set of social development programs,known as the missions. It also includes a host of standard questions thatallow scholars to gauge the socioeconomic, demographic, and partisanprofiles of respondents. Thus, the survey offers a unique opportunity togauge participatory democracy at the individual level using nationallyrepresentative data.

The results of the AmericasBarometer provide a mixed assessmentof the Bolivarian initiatives’ ability to incorporate and mobilize citizens.Most of the programs invite participation from less active segments ofsociety, particularly women, the poor, and the less educated; participa-tion in the Communal Councils and a few other programs is quite high.These are remarkable accomplishments that compare well with otherefforts in the region. However, much of this participation clusters withina narrow group of activists, and a disproportionate number of partici-pants are Chávez supporters, suggesting that these programs have a par-tisan bias.1 While less than ideal, these results do not indicate the use ofvote buying or other kinds of open conditionality, as some critics havesuggested (Corrales and Penfold 2007). Instead, they suggest thatVenezuelans tend to self-select into government programs for ideologi-cal reasons, particularly an affinity for the radical populist discourse andleftism of Chávez and his key allies. It is also likely that some of theseprograms generate real benefits that increase support from voters. Thus,the Venezuelan case suggests some optimism for proponents of partici-patory democracy, but also the need to be more attuned to its practical,political limits.

PARTICIPATORY DEMOCRACY

While there are multiple perspectives on the appropriate type and levelof participation in democracy (see Pateman 1970; Barber 1984; Santos2005), in this article, the term participatory democracy refers to the useof mass participation in political decisionmaking to complement or (inthe most radical versions) replace the traditional institutions of electionsand lobbying associated with representative democracy. As the Progres-sive era in the United States reminds us, participatory democracy is nota new concept. However, the emphasis on participatory democracy hasbecome an important component of leftist platforms over the past few

32 LATIN AMERICAN POLITICS AND SOCIETY 52: 3

decades, especially in Latin America (Roberts 1998; Avritzer 2002; Held2006, chap. 7).2 This emphasis constitutes an important shift from tradi-tional, Leninist approaches that treated democratic institutions as dan-gerous distractions from revolutionary goals; it celebrates democracy,and especially participatory democracy, as ends in and of themselves.

The participatory democratic emphasis of the New Left has acquiredadded relevance with the recent wave of leftist governments in LatinAmerica. Starting especially with their election victories in local areas,where they subsequently attempted many of these radical reforms(Chávez and Goldfrank 2004), leftist parties and movements have imple-mented a variety of participatory reforms across the region. The use ofparticipatory budgeting in Brazil has received most of the attention, butthe agenda includes other reforms, such the integration of new civilsociety organizations into traditional forms of corporatist policymakingand the implementation of direct primaries and more radical forms ofconsensual decisionmaking in political parties. Many of these reformsare now being endorsed by parties of the center and right, as well asmultilateral aid agencies, such as the World Bank, which see them inmore pragmatic terms as a means for providing better governance (seeShah 2007).

A growing number of scholarly works attempt to gauge the successof these reforms. Some studies focus on the policy outcomes of theseinitiatives, particularly their consequences for economic welfare and thequality of democratic governance. The results so far are mixed, show-ing that participatory reforms are implemented with much greater con-sistency and success at the local level than they are at the national level(Goldfrank 2008). In the emblematic case of participatory budgeting inBrazil, reforms have been found to have a host of positive conse-quences for the poor, including a reduction in clientelism (Santos 1998;Abers 2000; Serageldin et al. 2003; Navarro 2004; Wampler and Avritzer2004; for a review of this literature, see Goldfrank 2007).

However, most scholars who assess these initiatives tend to focuson the level and quality of political participation. The concern is to iden-tify whether significant numbers of previously excluded citizens arebeing mobilized and whether the programs that mobilize them repre-sent real instances of participatory democracy with meaningful budgets,autonomy, and popular involvement. The quantity and quality of par-ticipation are important for practical reasons, in that the incorporationof traditionally disenfranchised sectors, such as the poor, is presumablythe key to making government policy more representative and just, andespecially to eliminating the legacies of clientelism that typically char-acterize politics in the developing world. Yet participation is also impor-tant for its own sake, as a means of empowering citizens and givingthem control over their own lives.

HAWKINS: VENEZUELA’S DEMOCRACY 33

An important weakness of current studies of participatory demo-cratic initiatives is the lack of representative national samples of indi-vidual-level data. Often this is an artifact of the policy initiatives them-selves, which were conducted only in select municipalities and states.On this smaller scale, most analyses are case studies that focus onmovement participants or aggregate aspects of the participatory initia-tive (budget, results, decisionmaking procedures, etc.), not the largerpopulation, although small surveys of participants are sometimes avail-able (see Wampler 2007a, b; Goldfrank 2008). Hence, we have someidea of how many participate and what their participation means, butwe are not always certain about who these people are or how they com-pare with the rest of the population.

The study of participatory budgeting in Brazil exemplifies this lackof representative, individual-level data. Most of these analyses useaggregate municipal-level data. They find that participation tends to beconcentrated among Brazilians who already have experience as activists(Nylen 2002); that it seems to increase when programs receive effectivedecisionmaking power and are structured to allow better citizen delib-eration (Wampler and Avritzer 2004); and that it tends to be higher insmall municipalities that have higher levels of education and greatersupport for the Workers’ Party (PT) (Schneider and Goldfrank 2003).

All of these are valuable insights providing benchmarks for com-parison. However, they leave open the question of whether the attrib-utes they highlight also distinguish individual participants. We do notknow whether participants are also PT supporters and better educated,or whether support for the PT and higher levels of education across thecommunity ensure the creation of programs that subsequently incorpo-rate poor residents from other partisan backgrounds. A few studiesattempt to address this lack of data by using surveys of site-level par-ticipants (see especially Nylen 2002; Wampler 2007b), but these, in turn,raise questions about how well their results represent the entire com-munity. Ultimately, scholars need to complement these analyses withsurveys that sample from the entire population.3

PARTICIPATORY DEMOCRACY UNDER CHAVISMO

Chavismo in Venezuela is an extreme version of the push for participa-tory democracy by the left in Latin America. Before Chávez’s rise topower in 1998, previous governments had already experimented suc-cessfully with a series of basic initiatives to decentralize government,which included the direct election of local officials, a dramatic transferof central government revenues to states and municipalities, and thetentative implementation of internal party primaries for presidential can-didates (Mascareño 2000; De la Cruz 2004). New forms of civil society,


unallied with the traditional parties, proliferated (García 1992; Ellner1993; Buxton 2001).

Chavismo had mixed consequences for these reforms. Chávez andmany of his closest allies almost immediately sought to recentralize gov-ernment functions around his charismatic persona, ostensibly with thepurpose of carrying out their democratic revolution. Hence, local govern-ment allocations were increasingly controlled and manipulated by the pres-ident (Martín 2000; De la Cruz 2004), and secondary associations unalliedwith the government found themselves cut off from government commis-sions and financial support (Salamanca 2004). More recently, Chávez hastalked about replacing traditional local governments with a new form ofcommunity association, the Communal Councils, which would be tied tothe national government. And since the election of opposition mayors andgovernors in 2009, Chávez has worked to create supraregional governorswho would be directly appointed by the president.

Nevertheless, the government’s participatory rhetoric, its wholesaleproject of constitutional change, and its willingness to divert oil rev-enues to finance participatory initiatives are clearly without equal. Thenew constitution of 1999 makes participatory democracy a fundamentalprinciple of governance and citizenship, setting a high standard for thegovernment to meet in terms of level and quality of participation. Arti-cle 62 declares that “All citizens have the right to participate freely inpublic matters” and that “the participation of the people in the creation,execution, and control of public affairs is the required means to achievethe protagonism [protagonismo] that guarantees their complete devel-opment, both as individuals and as a collective.”4 Article 70 provides along list of government activities considered potentially participatory,including traditional ones, such as “elections to public office; the refer-endum; the consultation of public opinion; the recall of public officials;the legislative, constitutional, and constituent initiatives; the town hallmeeting; and the citizen assembly,” but also less traditional arenas, suchas “government offices open to the public; self-management; coman-agement; all forms of cooperatives, including those of financial nature;credit unions; [and] community businesses.” Constitutional reformsattempted in 2007 would have further expanded this list.

In an effort to achieve these ambitious goals, the government hascreated a rapid succession of quasi-partisan organizations and socialprograms that incorporate a heavy emphasis on local participation anddecisionmaking. Many of these Bolivarian initiatives have their originsin previous, non-Chavista efforts at grassroots mobilization in Venezuelaor existing government programs; however, all of them have receivedenormous impetus from the Chávez government and nowadays bear thestamp of the Bolivarian Revolution. The ones on which this studyfocuses are listed in table 1.5


Among the first to appear were the Bolivarian Circles, a vast net-work of voluntary associations that constituted the largest organizedcomponent of Chávez’s movement during his first years in power—comprising, at their peak, according to the government, approximately2.2 million Venezuelans. Called into action in 2001, each circle was toconsist of up to 11 members sworn to defend the constitution, be faith-ful to the ideals of Simón Bolívar, and serve the interests of their com-munity. They played a key role in mobilizing the popular response tothe attempted coup against Chávez in April 2002 and supplied campaignactivists during the presidential recall referendum of 2004, but subse-quently fell into decline (Hawkins and Hansen 2006).

At the height of the circles’ activity, the government also beganorganizing Health Committees (Comités de Salud), voluntary neighbor-hood committees created to help administer health clinics under theBarrio Adentro mission; they have essentially accompanied Barrio Aden-tro since its inception in 2003. Assuming that each clinic has one com-


Table 1. Key Participatory Initiatives of the Chávez Government

Year of Legal Number ofRecognition/ Participants (peak

Creation Purpose estimate by 2008)

Bolivarian 2001 Community improve- 2.2 million membersCircles ment and defense of

the Bolivarian revolution

Health 2003 Servicing of Barrio 6,500 committeesCommittees Adentro clinics

Cooperatives NAa Economic develop- 200,000 cooperativesment through alterna-tive forms of business enterprise

Urban Land 2003 Identity and legal 2–2.5 million membersCommittees entitlement of shanty-

towns

Communal 2005 Consolidation and 8 million participantsCouncils administration of

community develop-ment projects and municipal governance

aAs legally recognized entities, the cooperatives predate the Chávez government bymany years. Sources: García-Guadilla 2007a, b; Hawkins and Hansen 2006; López Maya 2008;MINCI 2007; SUNACOOP 2008.

mittee, we can safely estimate more than 6,500 Health Committees,although the exact number of individuals participating is unknown.6

Another important form of organization created by the governmentis the producer cooperatives. As of mid-2008, more than two hundredthousand of these (again, the exact number of participants is unknown)have been financed by government grants as part of its effort at“endogenous development,” which encourages economic diversificationand national self-sufficiency. The cooperatives depend on joint owner-ship and management but are still profitmaking enterprises (SUNA-COOP 2008). Also around 2003, the government began formally recog-nizing and encouraging the creation of Urban Land Committees(Comités de Tierra Urbana, or CTUs), shantytown associations that drafta community charter in order to receive legal recognition and title totheir lands. According to government estimates, more than 6,000 CTUshad been created as of 2007, incorporating roughly 2 million to 2.5 mil-lion adult Venezuelans (García-Guadilla 2007a).7

In 2005 the government began creating a vast network of Commu-nal Councils (Consejos Comunales). These are a new type of neighbor-hood association, tasked with combining and administering many of theabove entities. Each council is voluntarily constituted by up to fourhundred families in a given community, which meet in a CitizensAssembly (Asamblea de Ciudadanos); the council proper is an execu-tive committee selected by the assembly. The councils are not purelyterritorial, in that overlapping councils can exist in the same community.They represent the culmination of the government’s participatory dem-ocratic ambitions and constitute what it calls the “sixth branch of gov-ernment,” that of the “people or community.” Until the government’spackage of constitutional reforms was defeated in a December 2007national referendum, the councils were seen as potential replacementsfor traditional representative local government. Not only have theyreceived powers to coordinate many of the government’s other, morespecialized community programs, but they are eligible for sizable com-munity improvement grants formerly channeled to municipal and stategovernments, more than US$2.5 billion total in 2007.8 According to gov-ernment estimates, more than 33,000 councils had been organized bylate 2007, with more than 8 million Venezuelans participating (MPS2007; for more information on the Communal Councils, see García-Guadilla 2007b; López Maya 2008).

As forms of participatory democracy, many of these programs areproblematic. Most are dependent on government funding, and mustconform to regulations and oversight by the Chávez government, whichcan be intrusive and highly attuned to electoral objectives. Chávez’scharismatic authority is an omnipresent source of identity and motiva-tion among participants, and many of these organizations mimic (or


wholeheartedly endorse) the party line of the government. The Com-munal Councils in particular have come under heavy criticism for theirhasty implementation, their impractical design (they usurp governmentfunctions better handled at the municipal level), and their supervisoryrole over previously independent participatory initiatives in the com-munity, which seems to lead to government cooptation (García-Guadilla2007b; López Maya 2008; Lovera 2008). Yet these programs representinstances of local citizen debate and input requiring significant time andresources of participants; individual contributions are not directly com-pensated with any kind of salary (although financial benefits to the com-munity can be significant). Members are frequently allowed to developtheir own goals and everyday operating procedures and to choose theleadership of their organizations. And the programs are ostensibly tar-geted at populations that have traditionally been the most excludedfrom Venezuelan political life, such as the poor, the uneducated, shan-tytown dwellers, women, and those employed in the informal sector.

During the year before the presidential recall election of August2004, the government also began establishing a series of special povertyrelief programs called missions (misiones). The participatory element ismuch weaker in these programs, although there is important variancealong these lines, as the analysis will show. The missions are the cen-terpiece of the government’s unfolding program of socioeconomicdevelopment; they received several billion dollars in government oilrevenue between 2003 and 2005, as much as 3.5 percent of GDP,according to some estimates (Corrales and Penfold 2007). The full list ofmissions is quite extensive (Hawkins 2010; Hawkins et al. forthcoming);this study covers only the five oldest and best financed: Mercal, BarrioAdentro, and the three educational missions, known as Robinson II,Ribas, and Sucre. These are summarized in table 2.

Mercal is a series of subsidized supermarkets offering staple itemsat heavily discounted prices, sometimes as much as 40 percent; it con-sists of fixed locations (numbering in the hundreds) and street markets(numbering in the thousands), although the survey data examined heredo not distinguish among the types of stores. Barrio Adentro is a pro-gram of free health care, the backbone of which is several thousandsmall, primary care clinics. These originally were staffed by Cuban doc-tors, and today also include a more limited number of sophisticated clin-ics and hospitals. The educational missions provide remedial elementary(Robinson II) and secondary education (Ribas), as well as regular uni-versity education through a mixture of larger traditional campuses andcommunity college–like “municipal” programs (Sucre).9 According togovernment data (Campos 2006), by the end of 2005, Mission RobinsonII had nearly 1.5 million students, Mission Ribas had more than 600,000students, and Mission Sucre had about 250,000.


Unlike the Bolivarian initiatives, the missions are fully financed andformally responsible to cabinet-level agencies of the national govern-ment, and in some of them, associational life is minimal. In Barrio Aden-tro and Mercal particularly, aid workers are salaried employees, and aidrecipients are not required to make any significant effort to receive ben-efits or maintain the mission site. Therefore, it is important to draw a linebetween these programs and other, more clearly participatory initiatives.

The missions, however, provide a useful reference point for gaug-ing the reach of other government programs. Government leaders andactivists often see the missions as having a participatory component thatlinks them to civil society. The programs are bound up in the revolu-tionary ethos of the government, one that informs most workers, whoare frequently referred to as volunteers by cabinet-level officials and aidworkers (Hawkins 2010; Hawkins et al. forthcoming). They bringtogether and energize potential activists while providing resources forpolitical mobilization, such as education, a larger disposable income,and free time. Most of the missions incorporate some form of participa-tory governance, such as the Health Committees that accompany eachBarrio Adentro clinic. This sense of idealism and autonomy is especiallyhigh in the educational missions, which rely heavily on part-time work-ers, such as teachers from regular public schools and universities (noneof whom receive a stipend for their work in the missions), and whichrequire a much greater commitment of time and effort by participants.


Table 2. Key Social Missions of the Chávez Government

Date of Number ofPresidential ParticipantsDecree Purpose (date of estimate)

Barrio Adentro April 2003 Health care 12.5 million adults(2007)

Sucre July 2003 Decentralized 250,000 studentsuniversity education (2005)

Robinson II October 2003 Remedial primary 1.5 million education students (2005)

Ribas November 2003 Remedial secondary 600,000 studentseducation (2005)

Mercal January 2004 Subsidized food 8.7 million adults(2007)

Note: Estimates for Barrio Adentro and Mercal are not the number of visits but thenumber of unique users.Sources: Campos 2006; Hawkins 2010; Hawkins et al. forthcoming; author’s calcula-tions from 2007 AmericasBarometer survey.

These latter programs require students to draw up community develop-ment projects that give members of their communities a political voice,and they conscientiously introduce elements of participatory democracyinto the classroom experience. Because of this, the educational missionscan be included in the discussion of the Bolivarian participatory initia-tives here.

All of these efforts have begun to attract significant scholarly atten-tion. Increasing numbers of studies examine the performance of theBolivarian participatory initiatives and the more popular missions (Ellnerand Hellinger 2003; Corrales and Penfold 2007; Rodríguez 2007; Hsiehet al. 2007; Penfold-Becerra 2007; Ortega and Rodríguez 2008; Smildeand Hellinger forthcoming). However, much of what characterizes thebroader literature on participatory democracy also describes the studiesof Chavismo. None of these studies quantitatively assesses absolutelevels of participation and what sorts of cohorts are being mobilized.This is because they lack even the kind of survey data that are neededto assess questions of individual-level attributes across program sites, letalone across the country. A partial exception is López Maya’s 2008 studyof participatory democratic initiatives and self-help associations in theshantytowns of Caracas. In close studies of three particular sites, shefinds that numbers of participants as a percentage of their communityare very limited, and many of those who do participate are activists witha long trajectory of community involvement.

HOW MUCH PARTICIPATION

To determine how effective the Bolivarian initiatives are at mobilizingVenezuelans, we first need to know overall levels of participation. TheAmericasBarometer survey in Venezuela gauges popular participation inseveral types of associations and government programs with a nation-ally representative sample of 1,500 respondents. The survey includestwo modules of questions of particular interest. Specific wording andformat of all questions, as well as a more detailed description of thesample, are available by consulting the official report of the survey(Hawkins et al. 2008), although the question labels are provided herefor easier reference.

The first module (questions CP6 through CP17) measures the fre-quency of participation in secondary associations and participatory gov-ernment programs. For each organization, respondents are askedwhether they participate weekly, monthly, yearly, or never. The first halfof this module lists traditional forms of civil society and is the same inall versions of the AmericasBarometer in other countries. It includes reli-gious organizations, parent-teacher organizations, community associa-tions (specifically, “a committee or association for neighborhood


improvement”), professional/producer/peasant associations (hereafter,simply “producer associations”), labor unions, and political parties ormovements. In the second half of this module, respondents are askedabout their involvement in some of the most prominent Bolivarian par-ticipatory initiatives: the Health Committees, the cooperatives, the UrbanLand Committees, and the Communal Councils.

A second module of questions (questions MIS1 through MIS13) asksrespondents about their involvement in the missions. If respondents didnot use the services of a particular mission, the survey follows up withan open-ended question (with a fixed list of responses coded by theinterviewer) regarding why they chose not to participate. It is notablethat in the case of the educational missions, where classes typically meetseveral times a week, the survey asks only whether the respondent tookclasses, not how often.

Further, the survey asks participants a short set of questions(VENCP30 and VENCP31) regarding whether or not they ever belongedto a Bolivarian Circle. Because the remaining circles supposedly ceasedto exist in 2007 following the creation of the United Socialist Party ofVenezuela (Partido Socialista Unido de Venezuela, or PSUV), Chávez’snew government political party, the survey uses a two-part question thatallows analysts to distinguish among those who never participated,those who participated but no longer do so, and those who still con-sider themselves participants in a circle.

The results are generally positive. Figures 1 and 2 show levels andfrequency of participation in these different groups of participatoryassociations (both traditional and Bolivarian) and the missions. Thefigure for the circles indicates the total of all who report having ever par-ticipated, whether or not their circle is currently active. In the first figure,the activities with the greatest levels of participation are a few of themore traditional components of civil society: churches, parent-teacherorganizations, and neighborhood associations, in that order, whichattract annual participation of at least one-third and sometimes one-halfof Venezuelans. However, they also include one distinctly Bolivarian ini-tiative: the Communal Councils. Participation in the Communal Councils(35.5 percent) is exceptionally high, insofar as it reflects a relativelyrecent government initiative. If this figure truly represents the total adult(over 18) population, then at least 6 million Venezuelans have partici-pated in a Communal Council at some time over the past year. And ifwe extend this to the over-15 population (the Communal Councils,unlike the survey, are open to all Venezuelans over age 15), then wehave something quite close to the government’s estimate of 8 million.10

This level of participation is almost three times that of the olderBolivarian Circles (using either the survey-based estimates for the circlesor the government’s original estimates) and is much higher than that of


the participatory budgeting initiatives in Brazil, which is typically 2 to 8percent of the population in municipalities with the program andreaches as high as 10 percent only in the most successful areas(Wampler 2007a; Schneider and Goldfrank 2003).

A second set of organizations shown in figure 1 reveals relativelylower levels of participation. These are political parties, professionalassociations, and labor unions, in that order, as well as the Health Com-mittees, cooperatives, and Urban Land Committees. Again, extrapolatingthese numbers, they suggest that roughly 1.2 million (in the case oflabor unions) to 3.1 million adult Venezuelans (in the case of the HealthCommittees) participate at least annually in these associations. As withthe Communal Councils, the results for the Bolivarian initiatives showlevels of participation that are only slightly lower than the government’sestimates.

These levels are more typical of traditional civil society and partici-patory initiatives elsewhere in Latin America. Figure 3 displays analo-gous results for a few of these secondary associations for all countriesin the AmericasBarometer 2006–7. Levels of participation in these asso-ciations (political parties, producer associations, labor unions, and


Figure 1. Frequency of Participation in Traditional Civil Society andBolivarian Initiatives (as percent of population)

Source: AmericasBarometer 2007; author’s calculation using Stata SE 10.1

7.0

8.5

9.2

13.1

15.0

15.6

17.6

34.6

35.5

38.8

54.2

Labor Unions

Producer Associations

Urban Land Committees

Cooperatives

Political Parties

Bolivarian Circles

Health Committees

Neighborhood Improvement Assoc.

Communal Councils

Parent−Teacher Organizations

Religious Organizations

Weekly Monthly

Annually

neighborhood improvement associations) are usually lower in LatinAmerica than they are in the United States and Canada, but Venezuelais roughly in the middle of the pack in every instance. In the case of thegovernment’s participatory initiatives, we lack this same opportunity forprecise regional comparison, but even the least popular of the Bolivar-ian initiatives—the Urban Land Committees, with 9.2 percent of respon-dents representing perhaps 1.6 million adult Venezuelans—enjoys levelsof participation as high as those for participatory budgeting in Brazil, inpercentage terms.11 The most popular of these initiatives—the HealthCommittees—nearly doubles that level.

Looking at the missions (figure 2), we see a similar bifurcation. Therelatively nonparticipatory missions, Mercal and Barrio Adentro, haveextremely high levels of use, while the numbers of participants in theeducational missions are closer to what we find for the lower half oftraditional civil society and the Bolivarian initiatives. In fact, thenumber for Mercal is much higher than that of any other program: anextraordinary 71 percent of respondents indicate that they used its serv-ices at least once in the past year. The lower levels of participation inthe educational missions result partly from their narrow appeal: while


Figure 2. Frequency of Participation in the Missions (as percent of population)



4.44.5

5.66.9

7.88.1

9.310.0

11.213.4

15.015.715.9

16.918.118.819.3

20.521.121.621.7

35.5

ChileGuatemala

EcuadorCosta RicaEl Salvador

BrasilPanamá

ParaguayUruguay

MéxicoVenezuela

ColombiaNicaragua

PerúHonduras

United StatesBolivia

CanadaJamaicaGuyana

HaitíRepública Dominicana

6.87.07.58.18.59.19.910.110.110.410.5

12.112.8

14.915.1

16.619.420.0

23.228.5

39.141.2

ParaguayBrasil

El SalvadorChile

VenezuelaGuatemala

PanamáUruguay

HondurasCosta Rica

ColombiaMéxico

EcuadorRepública Dominicana

NicaraguaJamaica

HaitíGuyana

PerúBolivia

United StatesCanada

Figure 3. Frequency of Participation in Traditional SecondaryAssociations, by Country (as percent of population)

a. Political parties

b. Producer associations



c. Labor unions

d. Neighborhood improvement associations

1.3

2.0

2.2

3.3

3.8

3.8

4.6

4.7

5.9

6.1

6.1

6.9

7.0

7.3

7.8

8.1

8.8

10.3

14.5

El SalvadorParaguayHonduras

GuatemalaColombia

Costa RicaEcuadorPanamá

ChileJamaica

MéxicoNicaragua

VenezuelaHaití

BrasilRepública Dominicana

UruguayPerú

Guyana

14.414.7

18.221.121.922.422.4

23.523.8

25.227.728.3

31.533.0

34.534.734.935.1

39.143.2

44.747.6

BrasilUruguay

Costa RicaChile

PanamáParaguay

El SalvadorNicaraguaColombia

GuatemalaMéxico

JamaicaHonduras

HaitíRepública Dominicana

VenezuelaEcuadorGuyana

PerúBolivia

United StatesCanada

food and health care are more basic, consistent needs for most of thepopulation, remedial or higher education is required by only a smallsubset, particularly because most educational needs are already met byregular public and private education.12 Access to secondary and terti-ary education is moderately good for the region (gross enrollment in2000 was 59 percent and 28 percent, respectively), and for primaryeducation it is nearly universal (gross enrollment 102 percent; datafrom World Bank 2008).

Regarding the remaining Bolivarian association, the Bolivarian Cir-cles, the survey finds moderate levels of participation. Results indicatethat (again, extrapolating to the larger population) roughly 2 millionadult Venezuelans participated in a circle sometime in recent years. Thisis again close to the government’s estimates of 2.2 million members atthe peak of the circles’ activity in 2004. What is perhaps more interest-ing, and not shown in figures 1 and 2, is that half of these former par-ticipants—8.6 percent of Venezuelans—claim to be “currently partici-pating” despite the formal incorporation of the circles into the PSUV.This may reflect the wording of this question, which fails to systemati-cally ask all respondents if their circles are still active, or it may suggestsome unwillingness by hard core activists to let go of their organiza-tions. Either way, it shows that some of the circles have a surprisinglevel of institutionalization.

In sum, participation in most of the Chávez government’s participa-tory initiatives is moderately high, comparable to traditional voluntaryassociations in Venezuela and Latin America, and at least as high aslevels of involvement in participatory budgeting in Brazil. For the Com-munal Councils, levels of participation are quite high and comparableto those of the most popular (but least participatory) missions, BarrioAdentro and Mercal, as well as the most popular forms of traditionalcivil society. Most of these aggregate figures are only slightly below thegovernment’s estimates. While levels of participation obviously fall shortof the government’s stated objectives of a fully participatory democracythat incorporates all citizens, they are fairly impressive in cross-nationalperspective.

WHO PARTICIPATES AND WHY

While these figures tell something about overall levels of participationin the Bolivarian Revolution, they cannot show who is being mobilizedand whether this participation represents previously excluded segmentsof Venezuelan society. As a first step toward identifying who partici-pates, this study calculates whether there is any overlap among respon-dents. Do participants in some programs also tend to participate inothers, or is there relatively little crossover between memberships and


pools of aid recipients? This item is measured by running a factor analy-sis (principal components) using all of the above measures of frequencyof participation. Any strong factors that emerge are indications that agroup of associations or programs has an overlapping set of partici-pants. High factor loadings for any association or program (>.50) arehighlighted in table 3 for easy comparison.

The results in table 3 provide a somewhat more mixed assessmentthan the three figures. They indicate roughly five underlying dimensionsof participation and thus a more limited group of participants. The firstfactor captures almost all of the Bolivarian participatory initiatives,including the Communal Councils, the Health Committees, the UrbanLand Committees, and neighborhood associations. The inclusion of theneighborhood associations, which are not explicitly Bolivarian, maymean that respondents interpreted “committee or association for neigh-


Table 3. Factor Analysis of Frequency of Participation

Variable Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Uniqueness

Religiousorganization 0.03 0.01 0.01 0.08 0.82 0.32

Parent-teacherorganization 0.24 0.05 0.16 0.02 0.56 0.59

Neighborhoodassociation 0.76 0.09 0.14 0.07 0.13 0.37

Producer association 0.18 0.68 –0.11 –0.06 0.04 0.49Labor union 0.09 0.65 0.05 –0.02 0.01 0.57Political party 0.32 0.50 0.22 0.21 0.00 0.55Communal Council 0.79 0.12 0.20 0.04 0.05 0.31Health Committee 0.72 0.15 0.10 0.04 0.02 0.45Cooperative 0.34 0.50 0.04 0.05 0.00 0.63Urban Land

Committee 0.67 0.08 –0.01 0.17 –0.05 0.51Bolivarian Circle 0.05 0.44 0.45 0.27 0.06 0.52Barrio Adentro

Mission 0.20 0.07 0.74 0.09 0.05 0.39Mercal Mission 0.13 –0.05 0.81 0.02 0.02 0.33Robinson II Mission –0.04 0.20 0.04 0.69 0.21 0.44Ribas Mission 0.28 –0.17 0.05 0.70 0.06 0.40Sucre Mission 0.05 0.05 0.19 0.55 –0.32 0.56Eigenvalue 2.64 1.71 1.60 1.43 1.17Proportion explained 0.17 0.11 0.10 0.09 0.07

Notes: Principal component factor analysis with an orthogonal varimax rotation.Strong loadings (>.50) are highlighted. 70 parameters, 5 retained factors, n = 1,375.LR test: chi2(120) = 3692.9, p<0.0000. Source: AmericasBarometer 2007; author’s calculation using Stata SE 10.1.

borhood improvement” as a reference to the Communal Councils (thequestion about the councils appears later in the survey), or it may beevidence that participants in Venezuela’s older neighborhood associa-tions are a key component of these newer forms of Bolivarian partici-patory democracy; we cannot determine this merely by looking at thissurvey.

Still, the point is that the people who participate in one of the gov-ernment’s most participatory initiatives tend to participate in the otherparticipatory initiatives. Those who joined the Bolivarian Circles arelikely to be those currently involved in the Urban Land Committees andHealth Committees and are more likely to be involved in the Commu-nal Councils. These results persist even when we omit the CommunalCouncils (results not shown here). Given that all of these organizationsexcept the Communal Councils or neighborhood associations have sim-ilar numbers of participants (between about 9 and 18 percent of respon-dents participate at least annually, as we saw earlier), this suggests thepresence of a core group of activists giving vitality to the Chavista move-ment in Venezuela. At the same time, the fact that this group only partlyoverlaps with participants in traditional civil society (the next factor)shows that these programs are mobilizing new participants.

The second factor captures most of traditional civil society. Itincludes professional associations, labor unions, and political parties,none of which are exclusively Bolivarian and which often have long his-tories that predate Chavismo. While some of these associations havebecome highly partisan in recent years as part of the overall process ofpolitical polarization, opposition-led associations often confront newer,Chavista counterparts, resulting in an aggregate profile that appearsdeceptively neutral (see, e.g., the description of Venezuelan labor inEllner 2005). It is interesting that the cooperatives are included in thisdimension rather than the first one. As we will shortly see, this is pos-sibly because the cooperatives incorporate a group of participants thatis similar in many respects besides partisanship: they tend to be male,wealthier, and better educated. Overall, this second group of activists isprobably somewhat smaller than the first (between 7 and 15 percent ofVenezuelans).

The third and fourth factors correspond to the missions and neatlydivide them along the lines discussed earlier. That is, users of Mercal arealso likely to be users of Barrio Adentro (factor 3), while participants inany educational mission are likely to have participated in other educa-tional missions (factor 4). Mercal and Barrio Adentro, of course, drawon a much larger portion of the population and thus include many indi-viduals who may not use the educational missions or belong to any kindof civil society or participatory initiative. That both of these factors aredistinct from the factor capturing traditional civil society suggests that


participation in these programs is driven by a distinct set of sociode-mographic influences; this will be confirmed, especially for the case ofthe popular missions. Their lack of correlation with the Bolivarian ini-tiatives hints at the different purposes of the missions: they are, in theend, designed primarily to alleviate poverty and only secondarily tomobilize citizens.

Factor 5 seems to be a second dimension emphasizing traditionalcivil society, but a segment with a very different pool of participantsfrom that of the first one in factor 2. These are religious and parent-teacher organizations. As the analysis will show, these tend to drawfrom women and members of newer, non-Catholic religions that arenoted for higher levels of religious activity. Overall participation here isquite high.

An anomaly worth pointing out is that the Bolivarian Circles do notfit neatly into any one of these five factors. Instead, they fit loosely intotwo of them: factor 2 (traditional civil society) and factor 3 (the popularmissions). The overlap with factor 2 seems sensible: as Hawkins andHansen (2006) argue, the circles were more like civil society than a partof the state, and they may have initially drawn from an older set ofactivists; at the same time, the circles drew from a broad spectrum ofChávez supporters, which perhaps made them akin to the most popu-lar missions. Their failure to overlap more neatly with the first factor,Bolivarian participatory initiatives, is a puzzle that cannot be solved inthis study.

To better identify the profiles of individual participants, this studyran a series of multivariate statistical models of participation for each ofthe Bolivarian initiatives, the missions, and two of the most representa-tive forms of association in traditional civil society; namely, producerassociations (which covers the first set of traditional secondary associa-tions) and religious organizations (which captures the second set of tra-ditional secondary associations). To simplify the statistical models anddiscussion of the results, all response categories were recoded for theseprograms and associations to “0 = never participate” and “1 = participateat least yearly.” Therefore, high numbers indicate higher levels of par-ticipation. Because the resulting scale of participation is dichotomous,this analysis used probit rather than Organized Least Squares. It also ranmost of these models using the original four-point scale of responses(and hence, ordinal probit), but this yielded essentially identical resultsand made the interpretation more complicated, particularly because thesame analysis could not be run for the educational missions, all ofwhich have only two possible responses.

The explanatory variables are sex (0 = male, 1 = female; 49.9 per-cent of the sample is male), age (a continuous scale running from 18 to89 years; the average age is 36.3 years), income (an interval-level meas-


ure of household income using ten categories; the average at the timeof the survey was around Bs765,000), education (an interval-level meas-ure of the respondent’s level of formal education using 21 categories;the average was about the tenth grade), activism (a four-point ordinalscale measuring whether the respondent often tries to convince othersof political positions, recoded such that 4 = ”frequently,” 1 = ”never”;the average is closest to 2 = “rarely”), charisma (a four-point ordinalmeasure of whether “Chávez expresses a convincing vision of thefuture”), and Catholic (a dichotomous indicator of the respondent’sdeclared religion, where 1 = Catholic, 0 = other or none; 80.3 percentof respondents say they are Catholic).

Because partisanship is such an important concern, this variable isanalyzed more closely by breaking down charisma into a series ofdummy variables for each response, where Charisma4 corresponds tothe strongest level of faith in Chávez (“strongly agree,” 35.4 percent ofthe sample), Charisma3 is the next-lowest level (“somewhat agree,” 28.3percent), and Charisma2 is the next-to-lowest level (“somewhat dis-agree,” 10.2 percent). The lowest level of affect (“strongly disagree,”26.2 percent) is the reference category. For one of the models, that ofthe Ribas mission, the analysis also includes the squared value of edu-cation, education squared, to capture the nonlinear relationshipbetween education and participation in this particular program.13

Readers may wonder why this measure of confidence or faith inChávez, charisma, is used rather than a more standard partisan indica-tor, such as party identity or past vote choice. The first reason is that thenumbers of useful responses to both party ID and vote choice are ratherlow in the survey, and even among positive respondents the results areextremely lopsided. For example, only 33 percent of respondentsexpressed sympathy with any party, and of those, more than 70 percentmentioned either Chávez’s original party, Movimiento V República(MVR), or the newer PSUV. Likewise, while 77 percent indicate havingvoted in the presidential election of December 2006, nearly 76 percentof these respondents claim to have voted for Chávez. In the case of votechoice, these results partly reflect distortions found in all survey data,especially a survey taken almost a year after the election, but they leaverather low variance on both indicators. Charisma, by contrast, has anearly perfect response rate and a slightly more normal distribution thatstill correlates highly with vote choice (r = .66, p<.000).

A second reason for preferring charisma is that it is drawn from awidely used index of charismatic attachment from the organizationalbehavior literature, one that has been shown to have a nice ability tocapture powerful emotional attachment to political leaders (Merolla etal. 2007). In the current Venezuelan context, where so much politicalsympathy is driven by support for or opposition to Chávez, this seems


an apt substitute. In any case, using a more standard partisan indicator,such as vote choice, yields similar results.

The results of these models are found in table 4. Because of thelarge number of models, the table provides only the coefficients and anindicator of their statistical significance; coefficients are considered to bestatically significant only when p<.05 or less. What first stands out is theimportance of affect for Chávez for all forms of Bolivarian organizations,whether the participatory initiatives or the missions. Charisma4 in par-ticular is a strong predictor, significant at the p<.000 level across all ofthese programs. The relationship is in the direction we might expect,with stronger levels of affect for Chávez associated with higher levels ofparticipation.

In contrast, the two traditional civil society organizations (producerassociations and religious organizations) have no statistically significantrelationship to affect for Chávez; participants are equally for and againsthim. Instead, participation in these associations seems to be a productof standard sociodemographic factors. It should be noted that the resultsfor these two specific associations are broadly representative of theresults we would find if we considered other traditional associationsfrom these two categories. The most important determinants are the sexof the respondent (males participate much more often in producer asso-ciations, females in churches), education (highly educated respondentsparticipate more in churches), propensity to talk about politics (thosewith a higher propensity participate more in producer associations),and, to a lesser degree, age (older Venezuelans are somewhat morelikely to participate in both groups of traditional associations). Incomeis not a strong predictor, although it should be noted that this tends tobe highly correlated with education and is thus dampened in the model.Religious denomination would also be a good predictor of participationin churches, in that non-Catholics are much more likely to participateactively, but recoding this survey question as a dichotomous variable(annual participation in any church) means that we lose this correlation.

Most of these results are in line with the literature on civic partici-pation in other countries, such as the United States (see Verba et al.1995), which finds that participation usually increases with age, income,and education, particularly to the degree that these bolster citizens’political skills, and whenever citizens are particularly likely to feel astake or interest in the association. Thus, for example, Venezuelan menparticipate more in traditional forms of association that are linked tooccupation, while females participate more in associations linked to reli-gion and childrearing.

What factors distinguish among the different Bolivarian programs?The participatory initiatives (and this includes the educational missions)are driven by a particular set of additional variables that set them apart



Tab

le 4

. Res

ults

of

Pro

bit

Model

s of

Par

ticip

atio

n

Tra

diti

onal

Sec

ondar

y Ass

oci

atio

ns

Boliv

aria

n P

artic

ipat

ory

Initi

ativ

es__

____

____

____

____

____

___

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

Pro

duce

rRel

igio

us

Hea

lthU

rban

Lan

dCom

munal

Boliv

aria

nVar

iable

Ass

oci

atio

nO

rgan

izat

ion

Com

mitt

eeCom

mitt

eeCooper

ativ

eCounci

lCircl

e

Fem

ale

–0.4

1***

0.25

***

0.29

**0.

01–0

.03

0.23

**0.

11Age

0.01

*0.

01**

0.01

***

0.01

0.00

0.02

***

0.01

**In

com

e–0

.03

–0.0

1–0

.02

–0.0

1–0

.01

0.01

–0.0

6***

Educa

tion

0.02

0.04

***

0.01

–0.0

10.

03*

0.00

0.02

Educa

tion (

squar

ed)

Act

ivis

t–0

.15*

*0.

03–0

.13*

*–0

.07

–0.1

1*–0

.05

–0.2

6***

Char

ism

a4–0

.08

0.02

0.34

**0.

59**

*0.

34**

0.56

***

0.97

***

Char

ism

a3–0

.12

0.04

0.17

0.33

*0.

26*

0.31

**0.

38**

Char

ism

a20.

170.

080.

03–0

.33

–0.1

50.

12–0

.15

Cat

holic

0.08

–0.0

70.

170.

170.

140.

09–0

.02

const

ant

–0.6

6–0

.87*

*–1

.67*

**–1

.60*

**–1

.25*

**–1

.53*

**–0

.88*

*

N1,

187

1,18

91,

186

1,19

11,

187

1,18

41,

175

Chi2

35.6

630

.41

55.3

442

.82

29.7

580

.77

172.

3Pse

udo r

squar

ed0.

050.

020.

050.

050.

030.

050.

16Lo

g lik

elih

ood

–342

–805

–554

–372

–485

–747

–464


Tab

le 4

(co

ntin

ued

)

Mis

sions

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

Var

iable

Robin

son I

IRib

asSu

cre

Bar

rio A

den

tro

Mer

cal

Fem

ale

0.26

0.45

***

0.25

*0.

140.

16Age

0.01

–0.0

1–0

.01

0.01

***

0.00

Inco

me

–0.0

5*–0

.04

–0.0

5*–0

.03

–0.0

5*Educa

tion

–0.0

3*0.

22**

*0.

08**

*0.

00–0

.01

Educa

tion (

squar

ed)

–0.0

1***

Act

ivis

t–0

.09

–0.1

5**

–0.0

7–0

.05

0.03

Char

ism

a40.

79**

*0.

58**

*1.

22**

*1.

03**

*0.

96**

*Char

ism

a30.

49*

0.07

0.62

**0.

66**

*0.

84**

*Char

ism

a20.

230.

400.

380.

37**

0.44

**Cat

holic

–0.1

9–0

.47*

**–0

.22

–0.0

1–0

.04

const

ant

–1.6

1***

–1.8

4***

–2.4

9***

–0.7

2**

0.11

N1,

187

1,18

71,

173

1,17

51,

188

Chi2

47.7

293

.995

.96

150.

312

4Pse

udo r

squar

ed0.

100.

120.

150.

090.

09Lo

g lik

elih

ood

–217

–342

–279

–738

–610

Note

: N

um

ber

s ar

e unst

andar

diz

ed c

oef

fici

ents

*p<.0

5; *

*p<.0

1; *

**p<.0

01.

Sourc

e: A

mer

icas

Bar

om

eter

200

7, a

uth

or’s

calc

ula

tion u

sing

Stat

a SE

10.

1.

not only from the more popular missions but from traditional civil soci-ety. Specifically, participants in the Bolivarian initiatives are often morelikely to be female, although, as with participants in traditional civil soci-ety, they are also older and somewhat more likely to talk about politicswith their neighbors. Also, education and income matter less for partic-ipation, or if they matter they do so in a negative direction, where lowerincomes and education are associated with higher participation, the onepossible exception being the cooperatives. The educational missions inparticular are strongly associated with predictable educational patterns,based on their target populations.14

This overall profile of participation is similar to that of the subset oftraditional civil society represented by religious organizations. Theseresults reaffirm the hope that Bolivarian associations are mobilizing newVenezuelans, particularly sectors of the population, such as women andthe poor, that have traditionally been excluded from politics. In contrast,utilization of the more popular missions is primarily associated withaffect for Chávez and two other sociodemographic factors peculiar toeach mission: users of Mercal are more likely to be poor, and patientsat Barrio Adentro are more likely to be elderly. These latter outcomesmake sense, although, from an efficiency standpoint, it is disappointingto find that Barrio Adentro is not more carefully targeted at the poor.

Overall, the factor analysis and the multivariate regressions leavetwo broad findings regarding the scope and depth of Bolivarian popu-lar mobilization. The good news is that Bolivarian participatory initia-tives may actually be mobilizing previously unorganized Venezuelans.We cannot completely eliminate the possibility that newer Bolivarianprograms are simply winning over activists from traditional civil society,but the presence of several distinct dimensions—traditional civil societyand the Bolivarian initiatives and the missions—and the relatively mod-erate levels of continuing participation in traditional civil society sug-gests that participants in these groups of associations and programs aresomewhat different sets of people. In particular, the regressions suggestthat the Bolivarian initiatives and educational missions are better thansome parts of traditional civil society at reaching out to women, thepoor, and the less educated.

The bad news is that participation in the Bolivarian initiatives, likethat in traditional civil society or participatory initiatives in other coun-tries, is somewhat limited to a group of dedicated activists. The Com-munal Councils are a partial exception to this trend because of their rawnumbers, but many of the individuals who participate in a given gov-ernment participatory initiative are likely to be the same people partic-ipating in other government initiatives. This confirms the findings ofLópez Maya (2008). In addition, it seems clear that participation in theprograms of the Chávez government is strongly colored by partisanship.


Participants in both the missions and the participatory democracy initia-tives are much more likely to be supporters of Chávez than are non-participants, an association that is largely absent in instances of tradi-tional civil society.

EXPLAINING PARTISAN BIAS

This pattern of pro-Chávez activism across all the Bolivarian initiativesis not entirely surprising; it corroborates statistical analyses elsewhereshowing that allocations of government programs across states andmunicipalities in Venezuela (specifically, some of the missions) are asso-ciated with partisan criteria (Penfold-Becerra 2007; Hawkins et al. forth-coming). According to these analyses, program sites are distributed dis-proportionately to swing districts, where they are most likely to have animpact on undecided voters, and occasionally to core districts, wherethey may be used to reward supporters. Some critics of Chavismo arguethat these biases are symptoms of a larger problem of cronyism andclientelism (Corrales and Penfold 2007). According to this view,Chávez’s movement is rebuilding the political machines of Venezuela’straditional parties, whose tendency to condition government services onpartisan affiliation ultimately gave way to corruption and economicdecline.

While true open conditionality clearly occurs in some areas of theChávez government, this is probably not the best explanation for thesefindings.15 Rather, pro-Chávez bias is probably the product of a subtleprocess of self-screening driven by Chávez’s populist discourse and left-ist Bolivarian ideology. It may also reflect popular endorsement of gen-erous social programs in a context of strong economic growth, whichthe country experienced abundantly during the four years prior to thesurvey.

The latter of these two arguments is a relatively straightforwardapplication of retrospective economic voting theory (Lewis-Beck andPaldam 2000). According to this view, the high coefficients for charismaseen in the regressions are the result, rather than the cause of partici-pation: those who enjoy the benefits of these programs are afterwardmore inclined to view Chávez favorably. Unfortunately, it is impossibleto test this argument with the data at hand because it requires an instru-mental variable analysis; a wide array of indicators from the survey weretested, but none provided a useful instrument. The fact that Chavez’sapproval ratings rose dramatically after the inception of the missionsand other participatory initiatives (from about 40 percent to 60 percentduring 2003–4) suggests that retrospective evaluations are an importantpart of the story. At the same time, the strength of affect for Chávez israther uneven across different programs and is not always strongest in


those providing the most material benefits, making it likely that otherfactors are also coming into play.

The argument about ideological self-screening is less familiar andrequires a bit more explanation. Self-screening means a process of self-selection driven by largely ideological considerations. All programlocales contain symbols and powerful reminders of the government’srevolutionary program and incendiary discourse. A few programs, suchas the more popular, noneducational missions, demand little commit-ment to the government’s rhetoric; reminders of Chávez are a littleeasier to ignore for a few minutes of shopping or an occasional medicalexam. But Chavista discourse is unavoidable in the Bolivarian participa-tory initiatives and the educational missions, where the goals, materials,and activities are all suffused with the language of neo-Marxism,antiglobalization, and Chávez’s “socialism of the twenty-first century.”Participants must sit in meetings or classrooms for hours at a time,sometimes for several days each week, and interact in a very personalway with the leaders, instructors, and other participants. For non-Chav-istas, such partisan reminders can become unbearable.

While the retrospective economic voting argument cannot be testedhere, ideological self-screening can be discussed and compared toclaims about open conditionality. To begin with, assertions about openconditionality are simply not very credible, even in the case of the mis-sions. The partisan correlates of mission allocations that show up in sta-tistical analyses are hidden to the public and evident only after consid-ering large quantities of data; they cannot be a conscious influence onindividual choices to participate. After studying actual program sites, theauthor and his colleagues (Hawkins 2010; Hawkins et al. forthcoming)found that the missions do not seem to use any kind of overt partisanconditionality to determine eligibility. Both Mercal and Barrio Adentrohave a clear open-door policy requiring only that users wait their turnin line. Likewise, the educational missions do not openly use partisancriteria in distributing classroom slots or scholarships, nor do partici-pants suspect that there is any such screening.

The survey here seems to confirm these site-level findings. Thesurvey includes a question asking nonparticipants in the missions theirreasons for not using its services (MIS3, MIS6, MIS9, MIS12, MIS15).Nonparticipants almost never cite partisan concerns. In the cases ofMercal and Ribas, for example, fewer than 1 percent of nonparticipantssay they were denied services for political reasons, and only 1 to 3 per-cent mention their disagreement with the government’s politics. The evi-dence is not as strong for the participatory Bolivarian initiatives becausethe same follow-up questions were not asked, but the prominence andfunding of the missions lead one to suspect that this pattern holds acrossother, more participatory instances of popular mobilization, all of which


offer fewer benefits to participants and demand greater sacrifices of timeand resources. We should also bear in mind that one-quarter of the par-ticipants in most of the participatory initiatives actually have negativeaffect toward Chávez, a finding that seems to show that Chávez oppo-nents can and sometimes do participate in these programs, albeit in dis-proportionately lower numbers.16

We can see other evidence of ideological self-screening when welook more closely at the results of the probit analyses. The coefficientsfor charisma are large across all the government programs and espe-cially in the two most popular missions; Barrio Adentro and Mercalshow a very strong association with affect for Chávez. Yet the magni-tude of these coefficients is just as strong in Sucre mission and the Boli-varian Circles, both of which (especially the circles) offer more limitedbenefits and require greater sacrifices by participants. Partisan condi-tionality is probably not driving these latter results.

An additional piece of evidence comes as we consider not theabsolute magnitudes in the change in probability of participating inthese programs (which can be a little misleading in popular programswith higher baseline participation), but the percent increase in proba-bility. To gauge this, the expected probabilities of participating in eachof these participatory initiatives and the popular missions were calcu-lated, given an imaginary but fairly representative respondent with thefollowing characteristics: female, age 36, schooling to the 10th grade,7th income decile (Bs 680,000–750,000 per month for the household),and Catholic—and using all four possible responses to charisma. Thepercent change in probability across the different responses to charismawas then calculated.

As table 5 shows, the largest percent change is found not in theMercal and Barrio Adentro missions but in several of the participatoryinitiatives, especially the educational missions, the circles, and theUrban Land Committees. The probability of participating in these latterprograms is, of course, low, but much of the participation they do haveis explained by affect for Chávez. The change in probability from lowestto highest levels of affect (see the bottom row of the table) is usuallywell over 100 percent. This again seems to indicate strong self-selectioninto more ideologically suffused programs rather than open condition-ality or clientelistic vote buying. There is no reason for a clientelisticgovernment to impose higher partisan hurdles in programs providingthe smallest material benefits. However, there is a reason for highlydevoted followers of Chávez to select themselves into these programs,as they typically require the greatest level of long-term exposure andcommitment to the government’s discourse.

One final piece of evidence is found in the shape of these trends.Figure 4 depicts these expected probabilities graphically. The horizon-



Tab

le 5

. Exp

ecte

d P

robab

ility

of

Par

ticip

atin

g in

Boliv

aria

n P

rogr

ams,

by

Affec

t fo

r Cháv

ez

Res

ponse

to

Urb

anSu

rvey

Pro

duce

rCom

munity

Hea

lthLa

nd

Bar

rio

Robin

son

Boliv

aria

nQ

ues

tion

aAss

oc.

Counci

lCom

mitt

eeCooper

ativ

eCom

mitt

eeAden

tro

Mer

cal

IIRib

asSu

cre

Circl

e

Stro

ngl

y ag

ree

0.06

0.53

0.30

0.19

0.17

0.73

0.87

0.10

0.24

0.16

0.34

Stro

ngl

y dis

agre

e0.

070.

320.

190.

110.

060.

330.

570.

020.

100.

010.

08

Per

cent ch

ange

–13.

868

.254

.469

.117

9.9

117.

752

.542

2.0

141.

11,

114.

130

8.7

a Ques

tion: “C

háv

ez e

xpre

sses

a c

onvi

nci

ng

visi

on o

f th

e fu

ture

.”

tal axis measures the different levels of affect for Chávez, while the ver-tical axis gauges the likelihood of using each of these programs, usingthe results of the regressions in table 4. If some kind of partisan condi-tionality were in operation, we would see a jump in each line beyonda certain threshold of affect for Chávez; what we see, however, is theopposite. The graph shows some modest nonlinearities, but the twomost popular missions are flat or curve somewhat downward, suggest-ing a relatively low threshold of partisanship if there is one, while theeducational missions and the participatory Bolivarian initiatives tend tocurve upward, suggesting a higher threshold of partisanship. The effectis again highest in the Bolivarian Circles and the Sucre mission. Hence,if there is any partisan conditionality in effect, it is most severe in pro-grams that offer the smallest material benefits and require the greatestideological commitment.


Figure 4. Expected Probabilities of Participating in Bolivarian Programs, by Affect for Chávez


CONCLUSIONS

Participatory democracy remains an important issue distinguishing theagenda of the New Left in Latin America. However, data to analyze theeffects of participatory democratic initiatives on popular mobilizationhave been lacking, particularly in the salient case of Chavismo and theBolivarian Revolution in Venezuela. This analysis of results from theAmericasBarometer helps fill this gap.

What the case of Chavismo suggests for proponents of participatorydemocracy—and for the scholars who study it—is tempered enthusiasmand perhaps a little caution. Even in a country such as Venezuela, whichis dedicating enormous resources to participatory initiatives, the gains inpopular mobilization are real but moderate. Participation in the Com-munal Councils is very high, and most of the Bolivarian initiatives drawheavily from groups that are often absent from traditional civil society,especially women and the poor. These are encouraging results. Yetinvolvement in the most participatory Bolivarian initiatives seems to behighly concentrated among a group of activists. This finding necessarilyqualifies an assessment of the Chávez government’s efforts at greater cit-izen participation, and it suggests that broadly based participation willbe difficult to achieve in other countries as well.

The Bolivarian experiment also highlights a second area of concernthat receives relatively little attention in the broader literature on partic-ipatory democracy; namely, partisan bias. Involvement in all the Boli-varian initiatives is strongly associated with affect for Chávez. Whilesome of this support may be due to an ordinary, retrospective judgmentby voters receiving goods and services from the government, it proba-bly also represents the self-selection of participants based on their ide-ological affinity for Chavez’s populist discourse and leftist nationalism.This partisanship represents a potentially serious problem that under-cuts the government’s claims to operate participatory programs onbehalf of the Venezuelan people as a whole and confirms depictions ofVenezuela as a polarized polity (McCoy and Myers 2004).

It would be easy to dismiss this partisan bias as a problem uniqueto Chavismo in Venezuela, but it seems likely to occur in other LatinAmerican countries also. The most radical forms of participatory democ-racy are promoted most heavily by one side of the partisan spectrum,and these policies are rarely politically neutral in terms of the benefitsthey provide for different civil society organizations (see Goldfrank andSchneider 2006). The activist zeal that infuses these movements withenergy can alienate large swaths of the population that hold differingideologies and discourses—including many of the citizens who are sup-posed to be beneficiaries of these policies. Analysts of participatorydemocracy should be more attuned to this potential for bias in program


implementation and should include standard measures of partisanshipin their studies, at both the aggregate and individual levels.

NOTES

I would like to acknowledge the assistance and advice of Adam Anguiano,Jane Ann Fosson, Eliza Tanner Hawkins, several anonymous reviewers, and theparticipants in the Yale Conference on the Popular Sectors and the State inChávez’s Venezuela, including especially Susan Stokes, Thad Dunning, Fran-cisco Rodríguez, and Margarita López Maya. Special thanks also to MitchellSeligson and the staff at LAPOP for facilitating access to the AmericasBarometerdata. Funding was provided by generous grants from the Brigham Young Uni-versity College of Family, Home, and Social Sciences; the BYU David M.Kennedy Center for International and Area Studies; and the Jack E. and MarilynH. Roberts Family Endowed Fund in Political Science. The conclusions here,however, are my sole responsibility.

1. Partisan is used here in the broad sense of allegiance to a cause, fac-tion, or person, and not just to officially constituted political parties. Many sup-porters of Chávez remain mistrustful of parties per se and affiliate with thebroader movement by other organized means.

2. Avritzer (2002) distinguishes between what he calls participatory anddeliberative forms of democracy, but in the case of Chavismo as well as theBrazilian examples here, elements of both are often combined.

3. Again, in the case of Brazil, where participatory budgeting was imple-mented only in select localities, these surveys need not and probably should notinclude the entire country; the relevant population is the municipalities or stateswhere participatory budgeting has been implemented. But even locally repre-sentative survey samples are rare. For a partial exception in Brazil, see Wampler2007b.

4. All of these English translations are the author’s own. The original Span-ish text is widely available online.

5. Other organizations and programs include the Technical Water Round-tables (Mesas Técnicas de Agua), which work with government officials toreceive and apportion water in communities without regular service; the Self-Managing Community Organizations (Organizaciones Comunitarias Autóno-mas), shantytown associations designed to integrate community developmentprojects through a master plan; and the new community media outlets, whichprovide locally produced radio and television stations and newspapers. All ofthese incorporate strong elements of community deliberation and self-gover-nance. For more information, see López Maya 2008; García-Guadilla 2007; oncommunity media, see Fernandes forthcoming; Schiller forthcoming; Hawkins2006.

6. The government originally reported more than 10,000 Barrio Adentro Iclinics, but by November 2007 this figure had dropped to 6,527. See MINCI2007.

7. The Urban Land Committees predate the rise of Chávez and have tendedto demonstrate greater autonomy than some of the other initiatives; in this


regard, they are more a part of civil society than of the state. However, the gov-ernment has given them official backing and, since 2005, has attempted to givethem regular financing while bringing them under the aegis of the CommunalCouncils (García-Guadilla 2007a).

8. Specifically, the government reports as much as Bs5.2 trillion appropri-ated for the councils in 2007 (through November), with about 80 percent of thetotal actually distributed (MPS 2007). Some additional funding comes by way oflocal governments, a few of which are controlled by the opposition.

9. Perhaps the most famous educational mission is the literacy program,Robinson I. Because this mission is not a strong concern of the governmentsince it declared Venezuela “illteracy free” in 2005, the focus here is on thenewer phase of this program. (On Robinson I see Wagner 2005; for a criticalassessment of the actual success of its efforts, see Rodríguez 2007; Ortega andRodríguez 2008).

10. According to the National Institute of Statistics in Venezuela (INE 2008),in 2007 the adult (18+) population was estimated at 17.5 million, while the 15+population was 19.1 million. The latter yields an estimate of 6.8 million partici-pants in the Communal Councils, although we cannot be certain that the surveydata represent this younger cohort.

11. Bear in mind that these totals in Venezuela are for annual participa-tion, which in many of these programs is a rather low threshold for meaningfulinvolvement. Monthly and weekly participation are typically two-thirds of thistotal (figure 1).

12. That said, the educational missions actually show an increase fromwhat the government estimated in 2005. The number of participants is higherfor both Ribas (1.1 million, according to the survey, up from 0.6 million) andSucre (0.8 million, up from 0.25 million), but lower for Robinson II (0.6 million,down from 1.5 million). This may be due to the graduation of students fromlower programs, or it may represent the accumulation of nonperforming stu-dents in higher ones.

13. Participants in Ribas must have at least a grade school education butnot a complete secondary education. For consistency, all other models were runwith this same squared term, but none showed any effect greater than simplyusing the original, unsquared value.

14. Specifically, low levels of education are associated with participation inthe Robinson II mission, moderate levels of education with the Ribas mission(the inclusion of a squared term accounts for this quadratic relationship), andhigh levels of education with the Sucre mission (hence the positive sign of thiscoefficient). Non-Catholics are also more likely to use these educational mis-sions, particularly Ribas. This may reflect the fact that the best private schoolsin Venezuela are usually Catholic.

15. The most infamous case of open conditionality under the Chávez gov-ernment is the Lista Tascón, a list of all Venezuelans who signed electoral peti-tions for or against the government during 2003–4. The initial version of the listwas compiled by Luis Tascón, a Chavista politician, and remains widely distrib-uted in electronic form as a searchable database that identifies individuals byname, national ID number, and other personal information. During the run-upto the 2004 recall election, the list was used by the government to obstruct voter


registration and punish government employees suspected of supporting theopposition, and to a lesser degree by opposition businesses to punish their ownemployees who were government supporters. For a history of the list and ananalysis of its influence on the economy, see Hsieh et al. 2007.

16. This is true of all programs here except the Robinson II and Sucre mis-sions and the Urban Land Committees.

REFERENCES

Abers, Rebecca. 2000. Inventing Local Democracy. Boulder: Lynne Rienner.Avritzer, Leonardo. 2002. Democracy and the Public Sphere in Latin America.

Princeton: Princeton University Press.Barber, Benjamin R. 1984. Strong Democracy: Participatory Politics for a New

Age. Berkeley: University of California Press.Buxton, Julia. 2001. The Failure of Political Reform in Venezuela. Burlington:

Ashgate.Campos, Paulina. 2006. National Coordinator, Misión Sucre. Communication

with author. February 1.Chávez, Daniel, and Benjamin Goldfrank, eds. 2004. The Left in the City: Par-

ticipatory Local Governments in Latin America. London: Latin AmericanBureau.

Corrales, Javier, and Michael Penfold. 2007. Social Spending and Democracy:The Case of Hugo Chávez in Venezuela. LASA Forum 38, 1: 20–22.

De la Cruz, Rafael. 2004. Decentralization: Key to Understanding A ChangingNation. In The Unraveling of Representative Democracy in Venezuela, ed.Jennifer L. McCoy and David J. Myers. Baltimore: Johns Hopkins UniversityPress. 181–201.

Ellner, Steve. 1993. The Deepening of Democracy in a Crisis Setting: PoliticalReform and the Electoral Process in Venezuela. Journal of InteramericanStudies and World Affairs 35, 4 (Winter): 1–42.

–––––. 2005. The Emergence of a New Trade Unionism in Venezuela with Ves-tiges of the Past. Latin American Perspectives 32, 2: 51–71.

Ellner, Steve, and Daniel Hellinger, eds. 2003. Venezuelan Politics in the ChávezEra: Class, Polarization, and Conflict. Boulder: Lynne Rienner.

Fernandes, Sujatha. Forthcoming. Radio Bemba in an Age of Electronic Media:The Dynamics of Popular Communication in Chávez’s Venezuela. In Smildeand Hellinger forthcoming.

García, María Pilar. 1992. The Venezuelan Ecology Movement: Symbolic Effec-tiveness, Social Practices, and Political Strategies. In The Making of SocialMovements in Latin America: Identity, Strategy, and Democracy, ed. ArturoEscobar and Sonia E. Alvarez. Boulder: Westview Press. Chapter 9, 150–70.

García-Guadilla, María Pilar. 2007a. Ciudadanía y autonomía en las organiza-ciones sociales bolivarianas: los Comités de Tierra Urbana como movimien-tos sociales. Cuadernos del CENDES 24, 3: 47–73.

–––––. 2007b. El poder popular y la democracia participativa en Venezuela: losConsejos Comunales. Paper presented at the27th International Conferenceof the Latin American Studies Association, Montreal, September 5–8.


Goldfrank, Benjamin. 2007. Lessons from Latin America’s Experience with Par-ticipatory Budgeting. In Shah 2007. 91–126.

–––––. 2008. The Left and Participatory Democracy: Scaling Up or Scaling Back?Paper presented at the conference “Latin America’s Left Turn: Causes andImplications,” Weatherhead Center for International Affairs, Harvard Uni-versity, April 4–5.

Goldfrank, Benjamin, and Aaron Schneider. 2006. Competitive Institution Build-ing: The PT and Participatory Budgeting in Rio Grande do Sul. Latin Amer-ican Politics and Society 48, 3 (Fall): 1–32.

Hawkins, Eliza Tanner. 2006. Community Media in Venezuela. Paper presentedat the 26th Congress of the Latin American Studies Association, San Juan,Puerto Rico, March 15–18.

Hawkins, Kirk A. 2010. Venezuela’s Chavismo and Populism in ComparativePerspective. Cambridge: Cambridge University Press.

Hawkins, Kirk A., and David R. Hansen. 2006. Dependent Civil Society: TheBolivarian Circles in Venezuela. Latin American Research Review 41, 1:102–32.

Hawkins, Kirk A., Guillermo Rosas, and Michael E. Johnson. Forthcoming. TheMisiones of the Chávez Government. In Smilde and Hellinger forthcoming.

Hawkins, Kirk A., Jane Ann Patch, Adam Anguiano, and Mitchell A. Seligson.2008. Political Culture of Democracy in Venezuela, 2007. Nashville: LatinAmerican Public Opinion Project, Vanderbilt University. www.vanderbilt.edu/lapop

Held, David. 2006. Models of Democracy. 3rd ed. Stanford: Stanford UniversityPress.

Hsieh, Chiang-Tai, Edward Miguel, Daniel Ortega, and Francisco Rodríguez.2007. The Price of Political Opposition: Evidence from Venezuela’sMaisanta. Unpublished mss.

Instituto Nacional de Estadística (INE). 2008. Población total, según grupo deedad, 1990–2015. www.ine.gob.ve/poblacion/distribucion.asp

Lewis-Beck, Michael S., and Martin Paldam. 2000. Economic Voting: An Intro-duction. Electoral Studies 19: 113–21.

López Maya, Margarita. 2008. Caracas: The State, Popular Participation, and Howto Make Things Work. Paper presented at the Yale Conference “The Popu-lar Sectors and the State in Chávez’s Venezuela,” New Haven, March 6–8.

Lovera, Alberto. 2008. Los consejos comunales en Venezuela: ¿Democracia par-ticipativa o delegativa? Revista Venezolana de Economía y Ciencias Sociales14, 1: 107–24.

Martín, Guillermo. 2000. La República Bolivariana: ¿Relaciones interguberna-mentales en el siglo XXI venezolano? Perfiles Latinoamericanos 17 (Decem-ber): 123–51.

Mascareño, Carlos. 2000. Balance de la descentralización en Venezuela: logros,limitaciones y perspectivas. Caracas: Nueva Sociedad.

McCoy, Jennifer L., and David J. Myers, eds. 2004. The Unraveling of Represen-tative Democracy in Venezuela. Baltimore: Johns Hopkins University Press.

Merolla, Jennifer L., Jennifer M. Ramos, and Elizabeth J. Zechmeister. 2006.Crisis, Charisma, and Consequences: Evidence from the 2004 U.S. Presi-dential Election. Journal of Politics 69, 1: 30–42.


Ministerio del Poder Popular para la Comunicación y la Información (MINCI).2007. Inaugurados siete nuevos Centros de Diagnóstico, seis Salas de Reha-bilitación y dos Centros de Alta Tecnología. www.minci.gob.ve/noticias_prensa/28/16029/inaugurados_siete_nuevos.html

Ministerio del Poder Popular para la Participación y Protección Social (MPS).2007. 100 mil Consejos Comunales constituidos es la meta para el 2008.www.consejoscomunales.gob.ve/index.php?option=com_content&task=view&id=130&Itemid=73

Navarro, Zander. 2004. Participatory Budgeting in Porto Alegre, Brazil. In Lead-ership and Innovation in Subnational Government: Case Studies from LatinAmerica, ed. Tim Campbell and Harold Fuhr. Washington, DC: World BankInstitute.

Nylen, William R. 2002. Testing the Empowerment Thesis: The ParticipatoryBudget in Belo Horizonte and Betim, Brazil. Comparative Politics 34, 2:127–45.

Ortega, Daniel, and Francisco Rodríguez. 2008. Freed from Illiteracy? A CloserLook at Venezuela’s Misión Robinson Literacy Campaign. Economic Devel-opment and Cultural Change 57, 1: 1–30.

Pateman, Carole. 1970. Participation and Democratic Theory. New York: Cam-bridge University Press.

Penfold-Becerra, Michael. 2007. Clientelism and Social Funds; Evidence fromChávez’s Misiones. Latin American Politics and Society 49, 4 (Winter):63–84.

República Bolivariana de Venezuela. 2001. Líneas generales del plan de desar-rollo económico y social de la nación 2001–2007. Caracas. www.mpd.gob.ve/pdeysn/pdesn.pdf

Roberts, Kenneth M. 1998. Deepening Democracy? The Modern Left and SocialMovements in Chile and Peru. Stanford: Stanford University Press.

Rodríguez, Francisco. 2007. Sharing the Oil Wealth? Appraising the Effects ofVenezuela’s Social Programs. LASA Forum 38, 1: 22–25.

Salamanca, Luis. 2004. Civil Society: Late Bloomers. In McCoy and Myers 2004.93–114.

Santos, Boaventura de Sousa. 1998. Budgeting in Porto Alegre: Toward a Redis-tributive Democracy. Politics and Society 26, 4: 461–510.

Santos, Boaventura de Sousa, ed. 2005. Democratizing Democracy: Beyond theLiberal Democratic Canon. New York: Verso.

Schiller, Naomi. Forthcoming. Catia Sees You: Community Television, Clien-telism, and Everyday Statemaking in the Chávez Era. In Smilde andHellinger forthcoming.

Schneider, Aaron, and Benjamin Goldfrank. 2003. Budgets and Ballots in Brazil:Participatory Budgeting from the City to the State. IDS Working Paper 149.Brighton: Institute of Development Studies.

Serageldin, Mona, John Driscoll, Liz Meléndez San Miguel, Luis Valenzuela, Con-suelo Bravo, Elda Solloso, Clara Solá-Morales, and Thomas Watkin.2003.Assessment of Participatory Budgeting in Brazil. Cambridge/Washing-ton, DC: Center for Urban Development Studies, Harvard University/Inter-American Development Bank.

Shah, Anwar, ed. 2007. Participatory Budgeting. Washington, DC: World Bank.


Smilde, David, and Daniel Hellinger, eds. Forthcoming. Participation, Politics,and Culture in Venezuela’s Bolivarian Democracy. Durham: Duke Univer-sity Press.

Superintendencia Nacional de Cooperativas (SUNACOOP). 2008. Sunacoop haregistrado a 241 mil cooperativas a nivel nacional. www.sunacoop.gob.ve.

Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. Voice andEquality: Civic Voluntarism in American Politics. Cambridge: Harvard Uni-versity Press.

Wagner, Sarah. 2005. Venezuela: Illiteracy Free Territory. Venezuelanalysis, April21. www.venezuelanalysis.com

Wampler, Brian. 2007a. A Guide to Participatory Budgeting. In Shah 2007. 21–54. –––––. 2007b. Participatory Budgeting in Brazil: Contestation, Cooperation, and

Accountability. University Park: Pennsylvania State University Press.Wampler, Brian, and Leonardo Avritzer. 2004. Participatory Publics: Civil Society

and New Institutions in Democratic Brazil. Comparative Politics 36, 3:291–312.

World Bank. 2008. World Development Indicators 2007. CD-ROM. Washington,DC: World Bank.


Documents

Who Mobilizes? Participatory Democracy in Chávez’s ... · This article advances this discussion by studying popular mobiliza-tion under the government of Hugo Chávez in Venezuela,